The currently available tuberculin skin test (TST) and interferon gamma release assays (IGRA) cannot distinguish between infected persons who i) contain the infection and eliminate the organism (self-cure); ii) contain the organism but allow it to persist in a non-replicating state, providing a focus for future reactivation disease (truly latent infection); iii) contain the infection, which progresses to clinical disease within the first few years after infection (progressive infection). These three subpopulations of infected persons are grouped together as latent TB infection (LTBl). Biomarkers that differentiate among the three outcomes would allow targeting of preventive therapy to those most likely to benefit, facilitate development of new drugs active against specific stages of infection and promote durable cure. Predictive biomarkers are most needed for household contacts (HHC) of patients with infectious pulmonary TB (PTB), as they have the highest risk of developing active disease and, therefore, are an important focus for TB prevention programs. In Aim 1, we will employ FDG-PET/CT to characterize the early events in lymph node (LN) and lung following Mtb infection and apply the findings to guide the discovery of blood biomarkers predictive of high risk of progression to disease. In Aim 2 we will use multi-parametric flow technology and Cytokine Fingerprinting, a computational method for analysis of high-dimensional cytometric data, to derive a signature profile of effector and memory T cell subsets in treated HHC to discover biomarkers predictive of low risk of progression to disease and cure. The biomarkers of cure characterized in newly-infected individuals treated with isoniazid (INH) will guide identification of individuals that eradicate infection spontaneously (self-cure). Although memory T cells will be phenotypically similar in the drug-cured and self-cured individual, functionally they may differ since self-cured individuals achieved the status through immunity. In Aim 3, antigen-specific T cells from self-cured individuals will be identified by tetramers to discover correlates of protection. The proposed study will provide a new paradigm for TB prevention by facilitating specific treatments for populations at greatest risk for developing TB.